1. Field of the Invention:
The present invention relates to chip-on-boar encapsulation. More particularly, the present invention relates to an encapsulant molding technique used in chip-on-board encapsulation wherein an oxidizable metal layer is patterned on a carrier substrate and a top surface of the oxidizable metal layer is oxidized to facilitate removal of unwanted encapsulant material deposited during the molding operation.
2. State of the Art:
In the fabrication of semiconductor devices, a common circuit integration technique involves attaching individual semiconductor components, such as semiconductor chips, to a surface of a carrier substrate, such as a printed circuit board (e.g. FR-4), ceramic substrate, BT substrate, cyanate ester substrate, or silicon substrate, by any known chip-on-board attachment technique. Such chip-on-board attachment techniques include, but are not limited to, flip-chip attachment, TAB attachment, and wire bond attachment. After attachment, the semiconductor components may be encapsulated with a viscous liquid or gel insulative material (e.g., silicones, polyimides, epoxies, plastics, and the like) ("encapsulant material") with a transfer molding technique. This encapsulation (depending on its formulation) allows each semiconductor component to better withstand exposure to a wide variety of environmental conditions, such as moisture, ion impingements, heat, and abrasion.
An exemplary transfer molding technique for forming an encapsulant over a semiconductor component is illustrated in FIGS. 11-16. It should be understood that the figures presented in conjunction with this description are not meant to be actual views of any particular portion of an actual semiconductor component or molding device, but are merely idealized representations which are employed to more clearly and fully depict the process of the invention than would otherwise be possible.
FIG. 11 illustrates a pair of semiconductor components 202 attached to a carrier substrate 204 and in electrical communication with the carrier substrate 204 through a plurality of wire bonds 206. As shown in FIGS. 12 and 13, a multi-cavity encapsulant mold 208 is placed over the carrier substrate 204 and semiconductor components 202 (shown in shadow line in the top plan view illustrated in FIG. 12), such that cavities 210 (shown in shadow line in the top plan view illustrated in FIG. 12) of the multi-cavity encapsulant mold 208 are substantially centered over each semiconductor component 202. The multi-cavity encapsulant mold 208 is pressed against the carrier substrate 204 to prevent the border or other portions of the carrier substrate 204 from being covered by encapsulant material to be subsequently injected.
The cavities 210 of the multi-cavity transfer mold 208 are usually connected by an interconnection array of channels 212 connected to a central reservoir 214 (see FIG. 12) from which an encapsulant material, such as a molten particle-filled polymer, is fed under pressure. Usually, the channels 212 have constricted regions called "gates" 216 adjacent each cavity 210, as shown in FIG. 13. The gate 216 controls the flow and injection velocity of the encapsulant material 218 into each cavity 210 and forms a break point abutting the cavity 210 to permit removal of the excess channel encapsulant 222 which solidifies in the channels 212, as shown in FIG. 14. After the encapsulation of the semiconductor component 202 is complete and the encapsulant solidified, the multi-cavity transfer mold 208 is removed, as shown in FIG. 15. The excess channel encapsulant 222 at locations defined by channels 212 is then leveraged (shown in shadow lines in FIG. 15) from the surface of the carrier substrate 204 and broken free at an indentation 224 formed by the gate 216 (see FIGS. 13 and 14), called "gate break", as shown in FIG. 16.
The adhesion of the solidified encapsulant 218 to the carrier substrate 204 must be very strong such that the solidified encapsulant 218 does not detach from carrier substrate 204. However, this strong adhesion is disadvantageous when attempting to remove the excess channel encapsulant 222 from the carrier substrate 204. If the adhesion between the excess channel encapsulant 222 and the carrier substrate 204 exceeds the cohesive strength of the material of the carrier substrate 204 itself, the carrier substrate 204 will delaminate or rupture when the excess channel encapsulant 222 is leveraged from the surface of the carrier substrate 204.
Various methods have been devised to prevent the excess channel encapsulant from adhering to the carrier substrate. One such method is presented in U.S. Pat. No. 5,542,171 issued Aug. 6, 1996 to Juskey et al. ("the Juskey patent") which relates to treating a predetermined portion of the surface of the carrier substrate over which the mold channels will reside to prevent the excess encapsulant material thereon from adhering to the carrier substrate. The Juskey patent teaches selectively contaminating the surface portion with an ink or a polymer. A drawback of the Juskey patent is that applying inks or polymers to the carrier substrate surface risks contamination of the area adjacent a semiconductor chip, which contamination may prevent the adhesion of the encapsulant material over the semiconductor chip to the carrier substrate, resulting in a compromised package.
Furthermore, the technique taught in the Juskey patent would not be applicable to FR-4 substrates (flame retardant epoxy glass laminate). FR-4 requires a cleaning step, such as plasma cleaning, just before encapsulation to remove unwanted organic compounds in order to obtain sufficiently strong adhesion between the encapsulant material and the FR-4 substrate. Unfortunately, the plasma cleaning would also remove the ink or polymer as taught in the Juskey patent and, as mentioned above, addition of inks or polymers after such cleaning would risk contamination of the area adjacent a semiconductor chip. Thus, for an FR-4 substrate, the predetermined surface portion on the carrier substrate is plated with gold. The gold plating adheres to the FR-4 substrate, but not to most encapsulant materials. Also, this non-adhering property of the gold to encapsulant materials is not affected during the plasma cleaning of the carrier substrate. However, such gold plating is expensive.
An alternative arrangement of channels which injects the encapsulant material from the top (i.e., no excess encapsulant material on the carrier substrate when encapsulating the semiconductor component) has been used, but this requires a more complex and expensive molding system.
Thus, it can be appreciated that it would be advantageous to develop an inexpensive technique to treat a predetermined portion of the surface of the carrier substrate, over which the transfer mold channels will reside to prevent the excess encapsulant material from sticking to the carrier substrate while using commercially-available, widely-practiced semiconductor device transfer-molding packaging techniques.